Page 1

ISSN 2347 - 3983 Volume 1, No.1, September 2013

Journal of Emerging in Research, Engineering Research D Navya etInternational al., International Journal of Emerging Trends inTrends Engineering 1(1), September 2013, 16-20 Available Online at http://warse.org/pdfs/2013/ijeter04112013.pdf

Motion Vectors Used in Compressed Video D Navya 1 S. Naga Lakshmi2 D Keerthi 3 Dr Giridhar Akula4 1, 2,3

4

Students of Malla Reddy Institute of Engineering and Technology, Mysammaguda.India

Principal, Malla Reddy Institute of Engineering and Technology, Hyderabad4 seshagiridhar.a@gmail.com,navyadarba26@gmail.com

Abstract:-- In this paper we target the motion vectors

continuous event. Computers are digital systems;

used to encode and reconstruct both the forward predictive

they do not process images the way the human eye

(P)-frame and bidirectional (B)-frames in compressed

does.

video. The choice of candidate subset of these motion vectors are based on their associated macro block

2.PROBLEM STATEMENT

prediction error, which is different from the approaches

The main objective of this paper is

based on the motion vector attributes such as the magnitude

explained with four objectives, explained as follows.

and phase angle, etc. A greedy adaptive threshold is searched for every frame to achieve robustness while

First objective is to compile an introduction

maintaining a low prediction error level. The secret

to the subject of steganography. There exist a number

message bit stream is embedded in the least significant bit

of studies on various algorithms, but complete

of both components of the candidate motion vectors. The

treatments on a technical level are not as common.

method is implemented and tested for hiding data in natural

Material from papers, journals, and conference

sequences of multiple groups of pictures and the results are

proceedings are used that best describe the various

evaluated. The evaluation is based on two criteria:

parts.

minimum distortion to the reconstructed video and

The second objective is to search for

minimum overhead on the compressed video size. Based on

algorithms that can be used to implement for the

the aforementioned criteria, the proposed method is found

detection of steganographic techniques.

to perform well and is compared to a motion vector

The third objective is to evaluate their

attribute-based method from the literature.

performance. These properties were chosen because Key words: Least Bit significant, Hoffman code, stego

they have the greatest impact on the detection of

system, Tomography, DCT ,

steganography algorithms

1.

video

INTRODUCTION

In the below mentioned block diagram it has

This paper targets the internal dynamics of

been specified to hide data in motion vector of video

compression,

compression by using steganalytic system.

specifically

the

motion

estimation stage. Digital video refers to the capturing,

 The first is the looking scene that will hold

manipulation, and storage of moving images that can

the hidden information, called the cover

be displaced on computer screens. This requires that

image.

the moving images be digitally handled by the

 The second file is the message – the

computer. The word digital refers to a system based

information to be hidden

on discontinuous events, as opposed to analog, a 16


D Navya et al., International Journal of Emerging Trends in Engineering Research, 1(1), September 2013, 16-20

a format. This can be processed by subsequent steps. Preprocessing of the video is necessary to improve the detection of moving object’s For example, by spatial and temporal smoothing, snow as moving leaves on a tree, can be removed by morphological processing of the frames after the identification of the moving objects as shown in the following figure

3.NEW METHOD PROPOSED The secret message bit stream is embedded in the least significant bit of both components of the candidate motion vectors. By means of simple rules applied to the frame markers, we introduce certain level of robustness against frame drop, repeat and

Figure 1: Morphological Processing

insert attacks.

Another key issue in pre processing is the data format used by the particular background subtraction

 LSB Algorithm

algorithm. Most of the algorithm handles luminance

 Embedding data, which is to be hidden, into

intensity, which is one scalar value per each pixel, however, color image, in either RGB or HSV color

an image requires two files.

space, is becoming more popular in the background subtraction algorithms. Frame Separation: Image embedding method description: Frame processing is the first step in the background subtraction algorithm, the purpose of this

Take any input video, and then convert

step is to prepare the modified video frames by

that video in to number of frames. After converting

removing noise and unwanted object’s in the frame in

user need to select any one of the frame for

order to increase the amount of information gained

embedding the secret data. For hiding process we r

from the frame and the sensitivity of the algorithm.

going to use LSB technique. Embedded image is called as “Stego image” Then we are going to reconstruct that frames to videos.

Preprocessing is a process of collecting simple image processing tasks that change the raw input video info 17


D Navya et al., International Journal of Emerging Trends in Engineering Research, 1(1), September 2013, 16-20

Discrete Cosine Transform:

Least-Significant-Bit (LSB) Matching Method In order to keep the embedding of the same amount

DCT

coefficients

are

used

for

JPEG

of information as LSB matching and detect the secret

compression. It separates the image into parts of

data harder than the conventional LSB matching

differing importance. It transforms a signal or image

method,

from the spatial domain to the frequency domain. It

Mielikainen

proposed a

robust

LSB

matching method in 2006.

can separate the image into high, middle and low frequency

components.

Therefore, embedding message is performed for two pixels X and Y of a cover image at a time and then adjusting one pixel of the (X, Y) to embed two secret bits message s1s2.The embedding procedure is described as following:

Step 1. If the LSB of X is the same as s1, go to step Figure. 2 :Discrete Cosine Transform of an Image

2.

Procedure for DCT used in Steganography

Otherwise, go to step 3.

Image is broken into 8Ă—8 blocks of pixels. Working from left to right, top to bottom,

Step 2. If the value of f (X, Y) is the same as s2, do

the DCT is applied to each block.

not change any pixel. Otherwise, the value of pixel Y is increased or decreased by 1.

Each

block

quantization

Step 3. If the value of f (X −1, Y) is the same as s2,

is table

compressed to

scale

through the

DCT

coefficients and message is embedded in

the value of pixel X is decreased by 1. Otherwise, the

DCT coefficients.

value of pixel X is increased by 1. Where the function f (X, Y) is defined as Eq.1:

Procedure to Encode the Hidden Data

Since this new LSB matching method just only increase or decrease 1 in two adjacent pixels, the difference of the two neighborhood pixel between cover image and stego-image is very small. Hence, it can keep high quality while hiding data. 18


D Navya et al., International Journal of Emerging Trends in Engineering Research, 1(1), September 2013, 16-20

Take the DCT or wavelet transform of the cover image

2.

Frame separation of browsed video

Find the coefficients below a certain threshold

Replace these bits with bits to be hidden (can use LSB insertion)

Take the inverse transform

Store as regular image

To Decode the Hidden Data:

Total frames count= 80

3. Embedding data successfully and encryption time is 18.24 sec 

Take the transform of the modified image

Find the coefficients below a certain

Compressed file size = 42012

threshold 

Extract bits of data from these coefficients

Combine the bits into an actual message

I.

Experimental Results

1.

Input video by selecting browse video option

19


D Navya et al., International Journal of Emerging Trends in Engineering Research, 1(1), September 2013, 16-20

4. Decoding time is 11.27 sec and before decoding

show the electiveness of the proposed method. The

we have extract the embedded data. And it will save

results obtained also show significant improvement

in retrieve.txt file

than the existing method with respect to image quality and computational efficiency.

REFERENCES

[1] Ran-Zan Wang, Chi-Fang Lin, Ja-Chen Lin, Hiding data in images by optimal moderately signi7cant-bit replacement, IEE Electron. Lett. 36 (25) (2000) 2069–2070.

[2] A.Z. Tirkel, R.G. Van Schyndel, C.F. Osborne, A digital watermark, Proceedings of ICIP 1994, Austin 5. Here is reconstructed output is shown. If we

Convention Center, Austin, Texas, Vol. II, 1994, pp.

observe clearly the secret msg is exactly embed with

86–90.

that frame.

[3] W. Bender, N. Morimoto, A. Lu, Techniques for data hiding, IBM Syst. J. 35 (3/4) (1996) 313–336. [4] T.S. Chen, C.C. Chang, M.S. Hwang, A virtual image cryptosystem based upon vector quantization, IEEE Trans. Image Process. 7 (10) (1998) 1485– 1488. [5] L.M. Marvel, C.G. Boncelet, C.T. Retter, Spread spectrum image Steganography, IEEE Trans. Image Process. 8 (8) (1999) 1075–1083. [6] K.L. Chung, C.H. Shen, L.C. Chang, A novel SVD- and VQ-based image hiding scheme, Pattern Recognition Lett. 22 (9) (2001) 1051–1058. [7] Chi-Kwong Chan, L.M. Cheng, Improved hiding data in images by optimal moderately signi7cant-bit

4. CONCLUSION

replacement, IEE Electron. Lett. 37 (16) (2001) 1017–1018.

In this paper, a data hiding method by simple LSB

[8] Ran-Zan Wang, Chi-Fang Lin, Ja-Chen Lin,

substitution with an optimal pixel adjustment process

Image hiding by optimal LSB substitution and

is proposed. The image quality of the stego-image

genetic algorithm, Pattern Recognition 34 (3) (2001)

can

671–683.

be

greatly

improved

with

low

extra

computational complexity. Extensive experiments 20

Motion Vectors Used in Compressed Video  

In this paper we target the motion vectors used to encode and reconstruct both the forward predictive (P)-frame and bidirectional (B)-fram...

Read more
Read more
Similar to
Popular now
Just for you